Little progress has been made in the treatment of head-and-neck squamous cell carcinoma (HNSCC) in decades, and there are no impactful targeted therapies for this deadly cancer. The overexpression of the enzyme NQO1 has been observed in a high percentage of primary HNSCC tumors, including in those HNSCC patients with the worst prognoses, and thus compounds that are toxified by NQO1 have the potential to induce marked tumor regression for a desperate patient population. We have obtained a substantial amount of data showing the natural product deoxynyboquinone (DNQ) and its derivatives are potently cytotoxic in cells that express NQO1. As NQO1 is elevated in HNSCC but has low expression in normal tissue, DNQ and its derivatives have the potential as targeted drugs for HNSCC. In addition to significant in vitro and cell culture data on DNQ and derivatives, we have also made pharmacokinetic, toxicity, and efficacy assessments in mouse models of cancer and ? excitingly ? in pet cats with spontaneous HNSCC. Feline HNSCC is very similar to the human disease and is regarded as a significantly more representative and challenging pre-clinical model, thus our data showing activity of a DNQ derivative in these pet cats with spontaneous HNSCC in a NQO1-dependent fashion is very promising. DNQ and its derivatives we have identified thus far do not have the proper therapeutic index to become drugs, that is, their NQO1-independent toxicity limits the dose that can be given to mice and cats in vivo. We have experimentally ruled out various possibilities for the NQO1-independent cell death, and our experiments suggest that this toxicity is mediated by the reduction of DNQ by the one-electron reductase P450R. Herein we propose a comprehensive plan to widen the therapeutic index for DNQ and translate this drug class toward human clinical trials. In Specific Aim 1 we use a combination of structure-activity relationships and in silico modeling to design 80 novel DNQ derivatives that are predicted to be worse substrates for P450R without compromising their NQO1 activity. These compounds will be moved through a tiered series of cell culture, and patient-derived xenograft (Specific Aim 2) experiments, with the top compounds being evaluated in pet cats with HNSCC in Specific Aim 3. Our goal is to have identified a derivative suitable for translation to human clinical trials by the end of the funding period. This tightly-focused, hypothesis-driven proposal could provide the first impactful targeted therapy for HNSCC; this would be a major breakthrough for this vastly underserved patient population.